JP2003324742A - Image encoding apparatus and image decoding apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image encoding apparatus and an image decoding apparatus with less deterioration in image quality such as block distortion and mosquito noise even at a high compression rate. <P>SOLUTION: The image encoding apparatus includes: a first image encoding means for applying in-frame encoding to an input image according to a first prescribed encoding rule, sequentially outputting encoded information and outputting a reconfigured image of the encoded image; a second image encoding means for applying inter-frame encoding to the input image according to a second prescribed encoding rule, sequentially outputting encoded information and outputting a reconfigured image of the encoded image; an image encoding means for using at least either of the reconfigured image outputted from the first image encoding means and the reconfigured image outputted from the second image encoding means to perform inter-frame prediction encoding; and an encoded information composite means for composing the encoded information outputted from the first image encoding means with the encoded information outputted from the second image encoding means to output composed encoded information. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image encoding apparatus which encodes digital moving image data and outputs compression encoded information, and an image decoding apparatus which decodes compression encoded information and outputs digital moving image data. Regarding the chemical conversion device.

[0002]

2. Description of the Related Art Conventionally, many image compression coding systems have been devised for handling digital image data in broadcasting, communication, storage media and the like. Especially in these image compression coding methods, DCT (Discrete Cosine Transform)
A transform coding method in which orthogonal transform such as the above, quantization of orthogonal transform coefficient, and variable length code is combined is widely used.

In the transform coding method, the spatial frequency spectrum contained in the natural image has relatively few high frequency components,
Focusing on the fact that they are concentrated on the low frequency side, more bits are allocated to the low frequency coefficient than the high frequency coefficient among the conversion coefficients obtained by the orthogonal transformation. For example, as a standard mainly for still images, JPEG (Joint Photograp
hic Coding Experts Group) standard, but the initial standard JPEG (ITU-T T.81 | ISO / IEC10918-1) is DCT, DCT
It is an image compression coding method that combines coefficient quantization and variable length coding. JPEG is capable of compressing the amount of image data information to about one tenth while maintaining subjectively good image quality.

Further, in a moving image, since the correlation between adjacent frames is large, it is more efficient to encode the difference between adjacent frames than to encode the frames themselves. Such an encoding method is called predictive encoding.

However, in a portion where the screen movement is large, the correlation between simple frames becomes small, and the efficiency of predictive coding is lowered. Therefore, it is possible to further improve the coding efficiency by detecting the motion of the screen, compensating for the motion, and then performing the predictive coding. Such an encoding method is called motion compensation predictive encoding.

Motion-compensated predictive coding is particularly effective when the entire screen moves, such as when the camera is panned. Furthermore, a coding method called bidirectional predictive coding is also known. In bidirectional predictive coding, it is possible to further improve the coding efficiency by performing predictive coding from past and future neighboring frames.

[0007] For example, the MPEG (Moving Picture Experts Group) standard is a typical standard for moving images, but the MPEG standard is similar to JPEG in that it uses DCT / DCT coefficient quantization / variable length coding. Based on the transform coding method used, a bidirectional motion compensation prediction method is also used.

According to the MPEG standard, early MPEG1 (IS0 / IE
C11172-2), digital broadcasting and DVD (Digital Versatile
MPEG2 (ITU-T H.262 | ISO / used in Disk) etc.
IEC13818-2), MPEG4 (ISO / mainly used for communication etc.
There are types such as IEC14496-2).

The transform coding method using DCT or the like can efficiently compress an image, but when the compression rate is relatively high, image distortion called block distortion or mosquito noise occurs.

Since the amount of calculation in orthogonal transformation such as DCT increases exponentially with respect to the number of pixels, in the image compression coding method using orthogonal transformation, the entire screen is generally about 8 × 8 pixels called a block. , And the transform coding is performed in block units. Therefore, increase the compression rate,
When the quantization error of the transform coefficient becomes large, the DC component of each block becomes discontinuous, and a gap occurs at the block boundary. The gap between the blocks is visually recognized as block distortion.

Further, since the basis function of the orthogonal transformation is periodic, the presence of local edges in the block of the input image increases the high frequency components of the transformation coefficient. Due to the quantization error, a high frequency false signal is generated near the edge of the reproduced image. This false signal is visually recognized as mosquito noise. These image quality deteriorations significantly impair the subjective image quality, and improvements have been desired.

Due to such demands, recently, DWT (Discrete
Band-division filters such as te Wavelet Transform), quantization, and subband hierarchical coding methods using entropy coding are becoming popular. The sub-band hierarchical coding method divides an image into a plurality of sub-bands by a band-division filter, hierarchically repeats sub-band division for low-frequency sub-bands, and performs quantization and entropy coding for each hierarchical layer. It is a method of performing conversion. In this method, it is possible to improve coding efficiency by increasing the hierarchy of subbands to be divided.

In the sub-band hierarchical coding method, the amount of calculation is almost proportional to the number of pixels, and it is not necessary to divide the screen into blocks as in the transform coding method. When the bit allocation is finely controlled for each screen area, the screen is divided and encoded, but the entire screen can be encoded. In this case, the occurrence of block distortion can be suppressed in principle. Moreover, the basis function of DWT is aperiodic, and mosquito noise is unlikely to occur for a localized signal.

JPEG2000 (ITU-T
T.800 | ISO / IEC15444-1) is a subband hierarchical coding system that adopts DWT. In JPEG2000, instead of blocks, processing is performed for each screen area called tile. A tile is a rectangular area of an arbitrary size, and it is possible to divide the entire screen into a large number of small tiles and perform fine bit control for each tile. It is also possible to make the entire screen without dividing the tile into smaller parts. The image quality of JPEG2000 is superior to that of JPEG, from low compression rate to high compression rate.

JPEG and JPEG2000 are standards mainly for still images, but they can also be applied to moving images. Since a moving image can be regarded as a set of continuous still images, each image may be encoded in order. The method of encoding moving images using JPEG has not been officially standardized, but Motion-J
It is called PEG and is widely used in digital cameras. In addition, a method of encoding a moving image using JPEG2000 is commonly called Motion-JPEG2000, and JPEG2000 part3 (ITU-T
-T T.800 | ISO / IEC15444-3) is currently being standardized.

As an image coding apparatus and an image decoding apparatus using Motion-JPEG2000, for example, Japanese Patent Laid-Open No. 2001-2001
There is an image encoding device and an image decoding device described in 258031.

[0017]

By the way, in order to increase the compression rate while suppressing the occurrence of block distortion and mosquito noise in the compression encoding of moving images, a sub-band hierarchical encoding method and a motion compensation prediction method are used. It is considered effective to combine encoding methods. However, in the currently popular image compression coding standards such as the JPEG standard and the MPEG standard, there is no standard that adopts both the sub-band hierarchical coding system and the motion compensation predictive coding system.

Further, even if a coding method in which the sub-band hierarchical coding method and the motion compensation predictive coding method are combined is devised independently, the corresponding picture coding apparatus and picture decoding apparatus are not widespread. In addition, there is a problem that the application is limited.

It is an object of the present invention to provide an image coding apparatus and an image decoding apparatus which make good use of an existing image compression coding standard while achieving both a high compression rate and high image quality.

[0020]

[Means for Solving the Problems] In order to solve the above problems,
The image coding apparatus of the present invention performs intraframe coding on an input image according to a first predetermined coding standard, sequentially outputs coding information, and reconstructs a reconstructed image of the coded image. A first image encoding means for outputting and inter-frame predictive encoding for the input image according to a second predetermined encoding standard different from the first predetermined encoding standard, and encoding information is obtained. Second image encoding means for sequentially outputting and outputting a reconstructed image of the encoded image, wherein the reconstructed image output by the first image encoding means and the second image code Image encoding means for performing interframe predictive encoding using at least one of the reconstructed images output by the encoding means, encoding information output by the first image encoding means, and the second image Encoding information output by encoding means And encoding information synthesizing means for synthesizing and outputting a synthesis coding information, characterized in that it comprises a.
Further, the image decoding apparatus of the present invention separates the input synthetic coding information into coding information of the first predetermined coding standard and coding information of the second predetermined coding standard. Synthetic coded information separating means, and first image decoding means for decoding the coded information of the first predetermined coding standard output by the synthetic coded information separating means and sequentially outputting decoded images. A second image decoding means for decoding the coded information of the second predetermined coding standard output by the composite coded information separating means, and sequentially outputting decoded images,
A frame buffer for storing a decoded image output by the first image decoding unit and a decoded image output by the second image decoding unit, wherein the second image decoding unit is the first image; The inter-frame predictive decoding is performed using at least one of the decoded image output by the decoding means and the decoded image output by the second image decoding means, and the input synthetic coding information is , Composite encoding information in which information about the method of combining is embedded,
The composite encoding information separating means encodes the encoding information of the first predetermined encoding standard and the code of the second predetermined encoding standard according to the information on the combination embedded in the composite encoding information. It is characterized in that it is separated into digitized information.

[0021]

DETAILED DESCRIPTION OF THE INVENTION The present invention will be described below with reference to the drawings. In the following description, JPEG2000 is used as the first image coding method for intraframe coding, but the invention is not limited to this, and block distortion or mosquito noise does not occur such as a coding method using wavelet transform. Any image encoding method may be used.

Although MPEG2 is used as the second image coding method for performing interframe coding, it is not limited to this, and other interframe coding such as H.261, MPEG1 and MPEG4 is used. Any image encoding method can be used.

FIG. 1 is a diagram showing the configuration of a first embodiment of an image coding apparatus to which the present invention is applied. The image coding apparatus according to this embodiment includes a JPEG2000 coding unit 11, an MPEG2 coding unit 12, frame buffers 13 and 14, and a coding information synthesizing unit 15.

The frame buffer 13 temporarily stores the image data input to the image encoding device, and outputs the stored image data to the JPEG2000 encoding means 11 and the MPEG2 encoding means 12 in the order for bidirectional prediction. To do.

The JPEG2000 encoding means 11 encodes the image data output from the frame buffer 13 for each frame according to the JPEG2000 standard, and the generated JPEG2000 encoding information is supplied to the encoding information synthesizing means 15. In addition to the output, the reconstructed image of the encoded image is output to the frame buffer 14.

The MPEG2 coding means 12 performs intra-frame coding on the image data output from the frame buffer 13 according to the MPEG2 standard, or uses inter-frame prediction coding using reconstructed image data from the frame buffer 14. And outputs the generated MPEG2 coded information to the coded information synthesizing means 15 and the reconstructed image data of the coded image data to the frame buffer 14
Output to.

The frame buffer 14 stores the reconstructed image data output from the JPEG2000 coding means 11 and the MPEG2 coding means 12, and is referred to by the MPEG2 coding means 12 for performing interframe predictive coding. The image data is output to the MPEG2 encoding means 12.

The coded information synthesizing means 15 and the JPEG2000 coded information output from the JPEG2000 coding means 11 and MPEG2
The MPEG2 coded information output from the coding means 12 is combined and the combined coded information is output.

FIG. 2 is a diagram showing the structure of the JPEG2000 encoding means 11. The JPEG2000 encoding means 11 is the DWT means 2
1, a quantizer 22, an entropy encoder 23, an inverse quantizer 24, and an IDWT 25. JPEG2000
Since the configuration of the encoding means 11 is known, it will be briefly described below.

The image data output from the frame buffer 13 is input to the DWT means 21. In the DWT means 21, the image data is divided into tiles of a predetermined size,
The discrete wavelet transform is performed for each tile, and the DWT coefficient is output to the quantizing means 22.

The quantizing means 22 quantizes the DWT coefficients according to a predetermined quantizing table, and outputs the quantized coefficients to the entropy coding means 23 and the dequantizing means 24. In the entropy coding means 23, the quantized coefficient is coded by a predetermined coding table, and the generated coding information is sequentially output.

In the inverse quantizing means 24, the inverse operation of the quantizing means 22 is performed on the quantized coefficient and the inverse quantization is performed.
The DWT coefficient is output to the IDWT means 25. The IDWT means 25 performs an inverse discrete wavelet transform on the dequantized DWT coefficient, generates a reconstructed image, and outputs the reconstructed image to the frame buffer 14.

FIG. 3 is a diagram showing the structure of the MPEG2 encoding means 12. The MPEG2 coding means 12 includes a DCT means 31, a quantizing means 32, a variable length coding means 33, and an inverse quantizing means 3.
4, IDCT means 35, motion compensation prediction means 36, selector means 37, selector means 38, subtraction means 39, and addition means 310. The configuration of the MPEG2 encoding means 12 is also known, and will be briefly described below.

The MPEG2 coding means 12 operates differently depending on whether intra-frame coding is performed or inter-frame coding is performed. First, the case of performing intraframe coding will be described. The image data output from the frame buffer 13 is input to the selector means 37. Selector means 3
7 selects the image data from the frame buffer 13 and outputs it to the DCT means 31.

In the DCT means 31, the image data is 8 ×
The block is divided into blocks each having a size of 8 pixels, the discrete cosine transform is performed for each block, and the DCT coefficient is output to the quantizer 32. The quantizer 32 quantizes the DCT coefficient according to a predetermined quantization table, and outputs the quantized coefficient to the variable length encoder 33 and the inverse quantizer 34.

In the variable length coding means 33, the quantized coefficient is coded by a predetermined coding table, and the generated coding information is sequentially output. In the inverse quantization means 34,
The inverse operation of the quantization means 32 is performed on the quantized coefficient, and the dequantized DCT coefficient is output to the IDCT means 35. The IDCT means 35 performs inverse discrete cosine transform on the inversely quantized DCT coefficient to generate reconstructed image data, which is output to the selector means 38. The selector means 38 selects the reconstructed image data from the IDCT means 35,
Output to the frame buffer 14.

The case of performing interframe coding will be described. When interframe coding is performed, DCT means 3
1, the quantizing means 32, the variable length coding means 33, the inverse quantizing means 34, and the IDCT means 35 operate in the same manner as in the case of intra-frame coding, and therefore their explanations are omitted and the motion compensation prediction means 36 The operations of the selector means 37, the selector means 38, the subtraction means 39, and the addition means 310 will be described.

The image data output from the frame buffer 13 is input to the motion compensation prediction means 36. The motion compensation prediction means 36 compares the input image data with the reference image data output from the frame buffer 14, extracts the closest reference image data as a reference block, and outputs it to the origination means 39.

The subtracting means 39 subtracts the reference block from the block of image data to generate a prediction error and outputs it to the selector means 37. The selector means 37 selects the prediction error and outputs it to the DCT means 31. DCT means 31, quantization means 32, variable length coding means 33, dequantization means 34,
The IDCT means 35 operates on the prediction error in the same manner as in the case of intraframe coding.

The IDCT means 35 generates a reconstruction prediction error and outputs it to the adding means 310. The reconstruction prediction error output from the IDCT means 35 is added by the addition means 310 to the reference block output from the motion compensation prediction means 36, reconstructed image data is generated, and output to the selector means 38. The selector means 38 selects the reconstructed image data and outputs it to the frame buffer 14.

The operation of the frame buffers 13 and 14 will be described. As shown in FIG. 6, frame 0, frame 1, frame 2, frame 3, frame 4 of the input image
, Are encoded in the order of IBPBP.

Here, I represents image data which is intra-frame coded, P is unidirectional inter-frame predictive coding, and B is bi-directional inter-frame predictive coded. Since frame 1 is a B picture, bidirectional predictive coding is performed using frame 0 and frame 2. Therefore, it is necessary to encode frame 2 before frame 1. Similarly, frame 4 needs to be coded before frame 3 is coded.

Therefore, the frame buffer 13 outputs the image data in the order of frame 0, frame 2, frame 1, frame 4, frame 3 ... Since frame 2 is a P picture, when encoding frame 2, frame buffer 14 outputs the reconstructed image of frame 0 as reference image data.

Since frame 1 is a B picture, the frame buffer 14 outputs the reconstructed image data of frame 0 and frame 2 as reference image data when encoding frame 1. Similarly, the frame buffer 1
4 outputs frame 2, frame 2, frame 4, ... As reference image data. Note that the above description is an example, and there is no problem even if encoding is performed in other order / combination.

The operation of the coded information synthesizing means 15 will be described. FIG. 7 is an example of the MPEG2 coded information obtained for the input image data. SHC is a sequence layer code, GSC is a GOP (Group of picture) layer code, PSC0, PS
C1, PSC2, PSC3, PSC4 are the code of the picture layer corresponding to frame 0, frame 1, frame 2, frame 3, frame 4 ... M0, M1, M2, M3, M4 ... are frame 0, frame 1, frame 2, frame 3, frame 4
Are codes below the picture layer.

Here, the encoding information is represented in the order of M0, M2, M1, M4, M3, ... is not. By the way, in the inter-frame predictive coding, if the error is propagated due to the inter-frame prediction, the image quality is greatly deteriorated. Therefore, in order to suppress the propagation of an error due to inter-frame prediction, in general, the image quality of a frame serving as a reference picture is prioritized, and a large number of bits are assigned in the order of I picture, P picture, and B picture.

From this, the coding information synthesizing means 15 of this embodiment uses JPEG2000 coding information for the most important I picture and MPEG coding information for P picture and B picture. Therefore, since the frame 0 which is an I picture is used as the JPEG2000 encoding information, as shown in FIG. 8, the JPEG2000 for the frame 0 is set between the PSC0 and M0 as the user data start code and the user data UD0. Insert J0 which is encoding information. As a result, composite encoded information of the MPEG2 standard and the JPEG2000 standard can be obtained.

This coded information can be reproduced by a general MPEG2 image decoding device. In that case, JPEG20
J0 which is 00 coded information is not used and is treated as pure MPEG2 coded information.

This coded information is the MP for frame 0.
Since it contains both EG2 coded information and JPEG2000 coded information, the amount of information is slightly bulky. Therefore, as shown in FIG. 9, it is possible to remove M0 and instead insert only the user data start code and user data UD0, which is JPEG2000 encoded information J0 for frame 0. The synthesized encoded information obtained in this case has no waste of information amount, but cannot be reproduced by a general MPEG2 image decoding device.

In addition, the synthesis of the coded information shown above is
It can be applied to any frame. In this embodiment,
Although the JPEG2000 coding information is used only for the I picture, the JPEG2000 coding information may be used for the I picture and P picture.

By the way, since the block distortion and the mosquito noise are caused by the quantization error of the DCT coefficient as described above, they are more likely to occur as the quantization scale becomes larger. Therefore, by monitoring the quantization scale in the image coding device, block distortion and mosquito
The occurrence of noise can be predicted. Under conditions where block distortion and mosquito noise are likely to occur, JPEG20
By using 00 coded information and MPEG2 coded information in other cases, more efficient coding is possible.

FIG. 4 is a diagram showing the configuration of the second embodiment of the image coding apparatus to which the present invention is applied. This embodiment is different from the first embodiment in that a coded information composition control means 46 is added. The same parts as those in FIG. 1 are denoted by the same reference numerals as those in FIG. 1 and their explanations are omitted.

The MPEG2 coding means 12 outputs the quantization scale information to the coding information synthesis control means 46. The coding information synthesis control means 46 uses the quantization scale information output from the MPEG2 coding means 12 to determine the JP for each frame.
An encoded information synthesis control signal that specifies which of EG2000 encoded information and MPEG2 encoded information is used is generated and output to the encoded information synthesis means 15.

The method of generating the encoded information synthesis control signal is as follows.
For example, the average value of the quantization scale in the encoding target frame may be calculated and compared with a predetermined threshold value. If the average value of the quantization scale is larger than the threshold, JPEG2000
Encoding information is selected, and if the average value of the quantization scale is smaller than the threshold value, MPEG2 encoding information is selected. Further, the maximum value of the quantization scale may be used instead of the average value of the quantization scale.

The coded information synthesizing means 15, according to the coded information synthesizing control signal output from the coded information synthesizing control means 46, similarly to the first embodiment of the image coding apparatus of the present invention, the JPEG2000 code. The encoded information and the MPEG2 encoded information are combined and output.

According to the image coding apparatus of this embodiment, the JPEG2000 coding information is used only under the condition that block distortion or mosquito noise is likely to occur. As compared with the above embodiment, it is possible to efficiently suppress the deterioration of image quality.

FIG. 5 is a diagram showing the configuration of an embodiment of an image decoding apparatus to which the present invention has been applied. The image decoding apparatus according to this embodiment includes a JPEG2000 decoding unit 51, an MPEG2 decoding unit 52, a coded information separating unit 53, and a frame buffer 54.

The coded information is input to the coded information separating means 53. The coding information separating means 53 separates the inputted coding information into JPEG2000 coding information and MPEG2 coding information, and the JPEG2000 coding information is JPEG2000 decoding means 51.
Then, the MPEG2 encoded information is output to the MPEG2 decoding means 52.

The JPEG2000 decoding means 51 decodes the JPEG2000 encoded information output from the encoded information separating means 53 according to the JPEG2000 standard, and outputs the generated decoded image data to the frame buffer 54. MP
The EG2 encoding means 52 performs intra-frame decoding on the MPEG2 encoded information output from the encoded information separating means 53 according to the MPEG2 standard, or a frame buffer 54.
Inter-frame predictive decoding is performed using the reference image data from the frame buffer 54.
Output to.

The frame buffer 54 stores the decoded image data output from the JPEG2000 decoding means 51 and the MPEG2 decoding means 52, and MP the reference image data as necessary.
The decoded image data is output to the EG2 decoding means 52 and also in a predetermined reproduction order. With such a configuration, the image decoding apparatus according to the present embodiment can decode the combined coding information generated by the image coding apparatus according to the present invention.

[0061]

According to the present invention, it is possible to realize a moving picture coding apparatus and a decoding apparatus which have a high compression rate but which do not cause image distortion such as block distortion and mosquito noise.

[Brief description of drawings]

FIG. 1 is a block diagram of a first embodiment of an image coding apparatus according to the present invention.

FIG. 2 is a block diagram of JPEG2000 encoding means.

FIG. 3 is a block diagram of MPEG2 encoding means.

FIG. 4 is a block diagram of a second embodiment of the image encoding device of the present invention.

FIG. 5 is a block diagram of an embodiment of an image decoding apparatus of the present invention.

FIG. 6 is a diagram for explaining the operation of the frame buffer.

[Fig. 7] Fig. 7 is a diagram for describing MPEG2 encoded information.

FIG. 8 is a diagram for explaining a first example of combined encoded information.

FIG. 9 is a diagram for explaining a second example of combined encoded information.

[Explanation of symbols]

11 ... JPEG2000 encoding means, 12 ... MPEG2 encoding means,
13 ... Frame buffer, 15 ... Encoding information combining means.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5C059 KK03 KK04 MA00 MA04 MA05                       MA14 MA23 MA24 MC11 MC38                       ME01 PP05 PP06 PP07 RC12                       UA02 UA05 UA33                 5J064 AA01 BA09 BA16 BB01 BB03                       BC01 BC08 BC16 BC25

Claims (11)

[Claims] 1. Input image data is intra-frame encoded in accordance with a first encoding standard, encoded information is sequentially output, and reconstructed image data of the encoded image data is output. A first image coding means, inter-frame predictive coding according to a second coding standard for the image data, a second image coding means for sequentially outputting the coded information, the first Encoding information output from the image encoding means and encoding information output from the second image encoding means are combined, and encoding information combining means for outputting combined encoding information, 2. The image coding apparatus according to claim 1, wherein the image coding means performs interframe predictive coding using the reconstructed image output from the first image coding means. 2. Input image data is intra-frame encoded according to a first encoding standard, encoded information is sequentially output, and reconstructed image data of the encoded image data is output. A first image coding means, inter-frame predictive coding according to a second coding standard for the image data, sequentially output the coded information,
Second image encoding means for outputting reconstructed image data of encoded image data, encoding information output by the first image encoding means, and output of the second image encoding means And a coding information synthesizing means for synthesizing the coding information to output the synthesized coding information, wherein the second image coding means is a reconstructed image output by the first image coding means. An image coding apparatus, characterized in that interframe predictive coding is performed using at least one of data and reconstructed image data output by the second image coding means. 3. The encoding information synthesizing unit synthesizes the encoding information output by the first image encoding unit and the encoding information output by the second image encoding unit in frame units. The image coding apparatus according to any one of claims 1 and 2, characterized in that. 4. The encoding information combining control means, wherein the encoding information combining means outputs the encoding information output from the first image encoding means under the control of the encoding information combining control means. 4. The method of synthesizing the coding information output from the second image coding means and the method of synthesizing the information are embedded in the synthetic coding information. The image encoding device described in any one of them. 5. The coded information synthesizing control means controls the coded information synthesizing means using the quantization scale information output from the second image coding means. The image encoding device described. 6. The method according to claim 1, wherein the first coding standard is a standard using a wavelet transform.
The image encoding device according to 1. 7. The image coding apparatus according to claim 1, wherein the first coding standard is JPEG2000. 8. A composite coding information separating means for separating the composite coding information into coding information of a first coding standard and coding information of a second coding standard, and from the composite coding information separating means. A first image decoding unit that decodes the encoded information of the first encoding standard that is output and sequentially outputs the decoded image data; and the second image that is output from the synthetic encoded information separating unit. A second image decoding means for decoding the coded information of the coding standard and sequentially outputting the decoded image data; and the decoded image data output from the first image decoding means and the second image decoding means. A frame buffer for storing the decoded image data output by the image decoding means, wherein the second image decoding means outputs the decoded image data output by the first image decoding means and the second image decoding Decoded image data output by the means An image decoding device, characterized by performing interframe predictive decoding using at least one of the data. 9. The composite coding information includes information about a combining method embedded therein, and the composite coding information separating means stores the composite coding information according to the information about the combining method embedded in the composite coding information. The image decoding apparatus according to claim 8, wherein the image decoding apparatus separates the coding information of the first coding standard and the coding information of the second coding standard. 10. The image decoding apparatus according to claim 8, wherein the first coding standard is a standard using a wavelet transform. 11. The image decoding apparatus according to claim 8, wherein the first coding standard is JPEG2000.